Pharmacokinetics and disposition of memantine in the arterially perfused bovine eye

PURPOSE

To develop an improved ((1)This is to clearly acknowledge that we have tried to improve an existing model.) arterially perfused bovine eye model and investigate the general ocular disposition of memantine.

MATERIALS AND METHODS

Fresh bovine eyes were prepared by exposing and cannulating one ciliary artery, placing the eye into a perfusion chamber and slowly increasing the rate of perfusion to 1.0 ml/min. Analysis of the arterial perfusion pressure (APP), intraocular pressure (IOP), venous perfusate for glucose consumption and lactate dehydrogenase (LDH) activity, and histopathology ensured viability. Memantine was administered with the perfusate (simulated systemic access), by an intravitreal injection and by topical infusion. At the appropriate time points, the cornea, aqueous humour, sclera, iris-ciliary body, choroid/RPE, retina and vitreous humour were harvested and analysed for memantine.

RESULTS

The preparation remained viable for at least 9 h. At this time, histopathological examination showed mild to moderate deterioration of retinal layers. However, all retinal layers remained well defined and the integrity of the inner limiting membrane and Bruch's membrane were preserved. Glucose consumption, LDH levels and constant APP and IOP showed that correct cannulation and viability was maintained. After administration, memantine accumulated in the melanin rich iris-ciliary body and choroid/RPE. Results following topical administration indicate that substantial concentrations of memantine are present in the retina and choroid/RPE.

CONCLUSIONS

The arterial perfused bovine eye system proved to be a useful system for ocular drug delivery studies. The experimental results indicate that memantine will accumulate in the posterior segment when delivered by the topical route and that melanin-binding may support sustaining significant concentrations in the retina.

Binding of memantine to melanin: influence of type of melanin and characteristics

PURPOSE

The objectives of this study were to characterize sepia, synthetic, and bovine melanin and to determine their binding characteristics to the drug memantine.

METHODS

Physical methods were used to characterize sepia, synthetic, and bovine melanin. Their binding properties toward memantine were determined in deionized water and phosphate-buffered saline (PBS) at 37 degrees C. Melanin-memantine binding was measured indirectly by determining the unbound fraction of memantine. Curve fitting according to the Langmuir binding isotherm for one binding site was used for the determination of binding capacity (BLmax) and dissociation constant (KD).

RESULTS

Synthetic and sepia melanin had comparable Gaussian particle size distributions, whereas bovine melanin showed a heterogeneous distribution profile. The suspension medium had a small effect on the particle size distribution of synthetic and bovine melanin. There were characteristic differences in the infrared spectra of the melanins. The rank order for BLmax in deionized water was sepia > bovine > synthetic melanin. However, when the melanins were suspended in PBS, the BLmax values were lower, and the rank order was bovine > sepia > synthetic. Whereas the KD values for sepia and synthetic melanin remained largely the same in deionized water and PBS, the KD value for bovine melanin in PBS was more than twice than in deionized water.

CONCLUSIONS

This study showed that the physical characteristics of the melanins investigated differ markedly. The binding of memantine to melanin is thought to be determined by the different chemistries of the melanins, particle size, and buffer electrolytes.

Development of a liquid chromatography-mass spectrometric method for measuring the binding of memantine to different melanins

A sensitive and selective liquid chromatography-mass spectrometric method was validated for the determination of free memantine in melanin binding studies. The sources of melanin studied were sepia, synthetic and bovine melanin. Memantine was chromatographed on a reversed-phase column (Prodigy 5 microm, ODS(3), 100 A, 100 x 4.6 mm) using gradient elution with mobile phases of 0.1% formic acid in deionised water and 0.1% formic acid in methanol at a flow-rate of 0.8 ml/min. The mode of ionisation was atmospheric pressure-electrospray and detection by single ion monitoring of the memantine ion m/z 180. Validation of the method showed that the assay was linear from 0.1 to 1200 nM and 0.5 to 1200 nM memantine in deionised water and phosphate-buffered saline (PBS), respectively. Accuracy for sample preparations in deionised water was between 80 and 108% and between 80 and 123% for PBS. For both media, intra- and inter-day precision was below 1% for retention time and below 5% for analyte peak area. At the LLOQ, the variation of peak area was less than 17%. Binding of memantine to melanin was measured indirectly by determining the unbound fraction of memantine. After incubation of melanin with memantine, the sample was centrifuged and filtered to separate the memantine-melanin complex effectively from suspension. The filtrate was then assayed for free memantine from which the extent of binding was then calculated.